Protein drugs are one of the fastest growing segments of the pharmaceutical industry. The strong demand for these drugs reflects their ability to treat previously intractable diseases, including cancers, infectious disease, autoimmune disorders and cardiovascular disease. More than 40% of currently marketed protein drug products are amorphous solids, a form often chosen to prolong shelf-life and preserve potency. Nevertheless, protein drugs undergo a variety of physical and chemical degradation processes in the solid state. Aggregation is one of the most common of these processes. Since the presence of aggregates is associated with decreased potency and with an increased potential for life-threatening immunogenic side effects, they must be detected and removed during manufacturing and storage. This adds to the cost of producing protein drugs, ultimately increasing the cost to the patient and precluding the commercialization of promising new protein drugs that cannot be stabilized effectively. The goal of this research program is to develop rational methods for preventing protein aggregation in the solid state based on a thorough understanding of the chemical (i.e., covalent) and physical (i.e., non- covalent) mechanisms involved. The central hypothesis is that protein aggregation in amorphous solids is the result of specific covalent reactions and/or the exposure of aggregation-prone "hot spots" in the protein sequence, both of which can be prevented by designing the solid environment. Studies proposed for Specific Aim 1 will elucidate the mechanisms of thiol-disulfide exchange and disulfide scrambling in amorphous solids and will identify solid properties that control these reactions. The studies test the hypothesis that these common routes of covalent aggregation favor different pathways in solution and in the solid state and are influenced by solid composition. Specific Aim 2 will identify "hot spots" for non-covalent protein aggregation in amorphous solids using hydrogen/deuterium (H/D) exchange and molecular dynamics simulation (MDS). The work tests the hypothesis that these quantitative, high resolution measures of protein structure in amorphous solids will correlate with non-covalent aggregation during long-term storage. Specific Aim 3 will develop a computational model that predicts protein aggregation in amorphous solids based on properties of the protein and solid, producing a tool for formulation design and identifying variables critical to preventing aggregation. The work is relevant to the NIH mission of advancing the Nation's capacity to protect and improve health in that it addresses methods to preserve the potency and safety of a rapidly growing class of drugs. The work is also consistent with the agency's goal of ensuring a continued high return on the public investment in research by providing tools and knowledge for developing active proteins into marketable drug products. The presence of aggregates in protein drug products increases the potential for life-threatening immunogenic responses when the drugs are administered to patients. Understanding aggregate formation in amorphous solids will help ensure the safety of this rapidly growing drug class. The work will also help to control drug development costs by providing a rational basis for protein drug formulation.